5 ways AI will transform the future of utility asset management

Utility asset management is key.  It’s a big part of operations at every utility company.

First, regular inspections help ensure the safety and the reliability of your power grid. Second, they directly contribute to the efficiency of your grid operation.  They can deliver cost-effectiveness to your grid maintenance.  Ideally, you can maintain your infrastructure by detecting defects early on.  You can thereby minimize downtime.  You can optimize system reliability.  You can make maintenance proactive.  You no longer need to react to costly emergencies.

Typically, it takes 5 to 10 years to inspect all your distribution assets. You only evaluate about 10% to 20% of your distribution assets every year.  This makes it very difficult for you to maintain full situational awareness.  To be fully aware of the state of your grid.

As an innovation lead at a utility, you can now leverage the power of modern AI.  You have a transformational opportunity to solve utility asset management problems.  Inspections can happen more frequently.  You can do it at a fraction of the cost of traditional inspection.  You improve situational awareness multifold.  You can set the foundation for proactive planning.  You transform your ability to cater to increasing consumer demand.  You do it all while meeting your key performance indicators.

In this article, we share actionable insight to help you drive the change in utility asset management.

AI-powered utility asset management enables at-scale data collection and analysis

Autonomous data collection is an AI reality today.

You can enable at-scale data collection with more AI “eyes” on the ground.  You can integrate them into your regular workflows.  All this comes with little to no impact on your operations.  You can seamlessly integrate computer vision systems into your vehicle's fleet.

Here is how you can materialize this for utility asset management at your company.

Imagine your electric grid has 30 distribution assets per mile of roadway. Assume your fleet vehicles now integrate AI computer vision.  It autonomously collects data as the vehicles are on the road. This is simply part of routine operations. There is no operational change on that front.  The AI detects assets, geolocates them, and captures high-resolution imagery.  We can easily train AI models for you.  This way, you can analyze all images immediately.  You can see visible defects on your entire grid.

Let’s consider each vehicle drives 50 miles per day.  This means that per vehicle per day, you collect data for 50 x 30 = 1,500 assets. 

Now, let’s suppose you operate 100 service vehicles. Every day, you autonomously collect data on 100 x 1,500 = 150,000 assets.

Let’s go beyond this initial thinking.

Let’s look at what this means on an annual basis for your utility asset inspections.  You get millions of extra eyes on many more assets.  You do it without increasing operations and maintenance costs.  You significantly increase the amount of data collected.  With the power of AI, you can then use this data intelligently to conduct in-depth grid analyses.  You increase your company’s likelihood of finding critical defects. You address problem areas before they cause an outage.

As a head of innovation at a utility you can now understand how AI can transform operational workflows.  At-scale autonomous data collection and analysis forms the foundation to all of this.

In the following section, we’ll delve further into the importance of proactive maintenance.  We’ll also look into its cost benefits.

AI-assisted utility asset management enables proactive maintenance

Defect detection is a critical outcome of utility asset inspection and management.

For example, you can detect leaning poles, decaying wood, and corroded transformers. AI can transform the way you perform visual inspections today.

In 2022, we built a crossarm defect detection AI model for a customer that detected split and damaged wood on crossarms. 

On inspection routes, our cameras autonomously collect imagery and GIS data.  In just three months, our AI model flagged 1,646 assets with crossarm defects.  This was done automatically as the vehicle’s fleet drove by almost 16,000 assets on their grid.

However, this does not reflect the full picture of the crossarm defect detection model. This iteration did not categorize the severity of crossarm defects. With more defect imagery, we can improve the predictions of this model.  We can better account for the level of damage.  This also provides more value to our customers to help prioritize repair efforts. 

The more we collect data, the easier we make it to train your models.  Every year, we refine our models with more data coming in.  Soon, we will be able to provide a full suite of defect detection models.  One that is able to accurately identify relevant issues for individual assets.

AI-mediated utility asset management can help increases inspection efficiency.  Yet, it doesn’t always replace careful human inspection expertise. Rather, we see AI as an important tool to help flag assets.  These assets may necessitate further human validation.

With the help of AI, maintenance workflows at utilities become proactive as opposed to reactive.

It’s more focused on what matters.  It’s done faster.

To do that, proper resource allocation counts.  That’s what we are exploring next.

AI-driven utility asset management drives optimal resource allocation

Can you find new ways to prioritize grid maintenance work?

Imagine you could obtain comprehensive and high-quality data on your grid.  Data that accurately reflects its state. This is what AI-driven utility asset management can do for you.

Resource allocation is becoming increasingly complex for utilities. Climate change poses new threats to the aging grid infrastructure.  There are more severe weather events such as storms and wildfires every year. Utilities need to adopt new strategies for resource allocation.   This way, they can optimize budget dollars and avoid compromising on the frequency of utility asset management.

AI is here to help overcome this challenge.

AI-driven utility asset inspection is a low-cost, first pass review to find defects on the grid.  This traditionally is a mundane task.  It has a high cost. Now, thanks to AI, you can optimize maintenance and upgrade tasks.

As a result, you can shift resource allocation to more impactful capital work. You can upgrade your grid equipment instead of spending money on repairs. This is of significant importance.  AI helps you shift resources from O&M work to capital work. You can focus on increasing the grid reliability, resiliency and safety. 

This benefit goes beyond your resource optimization.

We’ll explore that in the next paragraph.

AI-enabled utility asset management allows data-driven decision making

Solid actionable insights come from extensive data collection.  

When you leverage precise and up-to-date grid information, you make better decisions.  

You can maximize your budget dollars with data-driven decisions.  You don’t solely lean on human experience.  You don’t uniquely rely on subjectivity.  You ground your decisions in hard numbers.

Autonomous at-scale data collection is big data.  It’s big actionable data.  It provides you with images.  It gives you asset coordinates.  And it gives you much more data.  Combine it with regular data collection intervals.  You can now track your grid condition over time. 

For example, you can visualize vegetation growth trends.  You can assess utility asset conditions over time.  You can pinpoint problem areas upfront.  You can optimize maintenance for grid impact.

You surface new patterns.  You create a new digital backbone for better decision making.

This new collaboration with AI transforms the way you decide the best course of action.

This gets better every year.

Your AI models improve as new data is collected and you solidify your new foundation for stronger decision-making.

This goes beyond just operation.

It expands into the future of your grid.  It applies to planning.  You create a resilient power future.  One that best weathers the risks of climate change and other threats to the grid.

Can you imagine the impact of machine learning on your grid in 2027?

By then, thousands of new computer vision techniques will be available to you.

How does this breakthrough AI innovation trend affect your ability to innovate?

AI for utility asset management is scalable and adaptable

The high-pace AI landscape will transform the way you conduct utility asset management.

Your AI learns fast.  You can train new models quickly.  You cover new use cases. The possibilities are endless.

Think lighting audits.  Think woodpecker holes.  Think idle transformers.  Think leaning poles. Think whatever your experience has in mind.

Today AI can learn it.  And it can learn it fast.  Whatever it is.

We can go beyond utility asset inspections through imagery.  We can enrich the data.  We can integrate other types of data. Point cloud or pseudo-LiDAR data can provide depth information to perform measurements between components for NESC clearance. Infrared cameras can capture thermographic data to identify hot spots.  Smart meters can detect anomalies.  It can preempt outage events. 

This way, AI can help you adapt.  You rise to constantly changing asset management requirements.  You change with changing environmental conditions.  You train new models.  You cater to specific rules or new limitations.

AI for utility asset management is also scalable. 

Current methods are slow.  They rely on human foot patrols.  The process is manual and costly. AI streamlines workflows to help make the grid safe. It makes it more reliable than ever before. 

Conclusion

AI can completely transform the way you conduct utility asset management. 

It leads to at-scale autonomous data collection and analysis.  It opens new predictive analytics. It creates new proactive maintenance possibilities.  It can help you drive optimal resource allocation.  Drive the shift from operational expense to capital expense for growth.  It allows data-driven decision making.  It expands it beyond solely utility operation and maintenance.  It constantly adapts. It always scales.  You can train new AI models.  You move faster.  You scale and expand into a more sustainable and resilient grid.  You innovate and increase profits.

Heads of innovation at electrical utilities—want to learn more about how Noteworthy AI can save your field operations time and money with AI?  

Drop us a line here for a free consultation to explore a low cost, low risk AI-powered utility asset management pilot.